View PDF - ijournals

A Web Service Quality of Protection using
AHP Method
V.Prasath
Assistant Professor, PKIET, Karaikal, U.T.Puducherry
ABSTRACT
As the number of Web services on the internet increase,
the need for finding the exact web service that matches
the user’s request also increases. So ranking of web
services is required in order to find the right web
service. The main goal of this research paper is to
achieve security of the web service can be summarized
to this single value. In this paper we propose analytical
hierarchy process (AHP) to evaluate the weights of
criteria instead of collect the weights directly from
service consumer. Next to reduce the complexity of
involving all the security parameters we follow the
greedy approach to evaluate and estimate each and
every parameter of web service security.
General Terms
Web Service Security
Keywords
Ws-Security, AHP, Greedy, Service Discovery
1. INTRODUCTION
Web services are one of the most promising
technologies for building distributed systems that has
the potential of becoming the core of a new webbased
middleware
platform,
providing
interoperability between computational services. In
this specific context security is very important
feature.
Nowadays,
many
companies
and
organizations implement their core business and
application services over Internet. Thus, the ability to
efficiently and effectively select and integrate interorganizational and heterogeneous services on the
Web at runtime is an important step towards the
development of the web service applications [1]. If
multiple Web services provide the same functionality,
then a Quality of service requirement can be used as a
secondary criterion for service selection. Next high
importance of Web services is security. To ensuring
the web service to some level of security to real
systems has been evident since it has been discovered
that most attacks may exploit vulnerabilities [2, 3, 4].
These vulnerabilities stem from web service are
poorly designed and developed. Security is a set of
non-functional
attributes
like
confidentiality,
integrity, availability. The current Universal
Description, Discovery and Integration (UDDI)
registries only support Web services discovery based
on the functional aspects of services. To practically
examine the resistance of attacks of a small subset of
security patterns that are commonly used in web
© 2014, IJOURNALS All Rights Reserved
service applications. To perform this evaluation, it has
to built a system with web services security testing
patterns and using them to study systems under
known categories of attacks to web service
applications [5] and determine which aspects of
security are enhanced through the use of each
security pattern for web service system. This paper
will describe a testing methodology for web services
security and outline a process that can be adopted to
evaluate web services security can be summarized to
single value. To define security as a measure of
vulnerabilities in the accuracy of a risk or security
measurement, analytical hierarchy process is used. In
order to render the definition useable, it is necessary
to associate the terms in the definition with a
measurement scale that represents the security of a
system as a value between 1 (secure) and 9
(insecure).
2. RELATED WORK
Several authors conducted performance evaluation
studies of WS-Security. Security for web services
means providing authentication, authorization,
confidentiality, and non-repudiation as basic
representative [6]. Each of these aspects is described
below
Li Jiang, Hao Chen, Fei Deng, Qiusheng Zhong [7]
proposed Web service security evaluation method
based on threat classification is proposed, which can
process security evaluation to Web service from
different angles of view, such as spoofing, tampering,
repudiation, message disclosure, denial of service and
elevation of privilege, and can provide a referential
evaluation index of Web service security for the users
through the threat modeling and evaluating the
degree of security.
Bachar Alrouh and Gheorghita Ghinea [8] consider
and compare the performance of various security
mechanisms applied on a simple web service tested
with different initial message sizes. The test results
show that transport layer security mechanisms are
considerably faster than message level security
mechanisms.
Artsiom Yautsiukhin. [9] argue that security must be
guaranteed for data processing and requirements
must be negotiated with a client and inserted into the
agreement between a client and a contractor. The
problem is that a client and a contractor have
different views on how these requirements should
look like. We propose a methodology which binds
Page 93
these views and describes a process for selection the
security configuration that helps to achieve
negotiated level of protection.
GannaFrankova and ArtsiomYautsiukhinork [10]
propose a methodology that identifies the concrete
business process providing the highest quality of
service and protection among all possible design
alternatives. the orchestrator should design its
business process aggregating web services in such a
way as to make it more efficient from the quality of
service and protection point of view.
González, Ricardo Mendoza [11] propose in a
theoretical manner including the knowledge offered
by the proposed specification based on patterns in a
new level for the architectural structure of WSSecurity specification, which is currently one of the
most popular specifications to establish secure web
services.
Duan Youxiang , Gao Yang. [12] says vulnerabilities
have been threatening Web Services confidentiality,
so it is meaningful to quantitatively evaluate them,
which can reflects web services security reasonably
and directly, and it will be convenient for user to
choose service and deploy security measures.
AHP(Analytic Hierarchy Process) algorithm is been
involved in this paper to dissect indicator. Operations
research and fuzzy mathematical theory are also used
to qualitatively and quantitatively evaluate
vulnerabilities based on the rank of web services
confidentiality.
3. PROPOSED WORK
In this proposed approach as shown in Figure 1, first
use of greedy approach to select desired security
parameter and the weighting of each criteria will be
evaluated by using AHP and then ranking the web
services based on this evaluated value. The steps of
our approach are shown in below
Recognize
Relation of
Criteria
criteria and the options provided by the user. In
addition, the AHP is simple because there is no need
of building a complex expert system with the decision
maker’s knowledge embedded in it.
AHP is a pair wise comparison method that each
criterion is comparing to each other and gets the
score with respect to Table 1.The AHP can be
implemented in three simple consecutive steps:
1) Computing the vector of criteria weights
2) Computing the matrix of option scores
3) Ranking the options
Table 1. Preference Level
Descriptor
Value
Exceptional
Outstanding
Excellent
Very Good
Good
Satisfactory
Fair
Marginal
Poor
3.2 Computing criteria weights and
Pair Wise matrix
In the following the steps of AHP pair wise evaluation
are described:
Step1: Sum all the values in each column.
Step2: The values in each column are divided by the
corresponding column sums.
Step3: Convert fractions to decimals and find the
average of each row. This sum is corresponding to
weight of the criterion of the row.
A
B
C
A
1
x
1/y
B
1/x
1
1/z
C
y
z
1
Greedy
AHP
Weights of
Criteria
Numerical
Score
1
2
3
4
5
6
7
8
9
Fig 2. Evaluation Weight Matrix
Ranking of
Web Services
Fig 1: Quality of Protection using AHP
3.1 AHP
The AHP is a very flexible and powerful tool because
the scores, and the final ranking, are obtained on the
basis of the pair wise relative evaluations of both the
© 2014, IJOURNALS All Rights Reserved
In order to compute the weights for different criteria,
the AHP starts creating a pair wise comparison matrix
A. The matrix A is a m*m real matrix, where m is the
number of evaluation criteria considered each entry
ajk of the matrix A represents 1the importance of the
jth criterion relative to the kth criterion. if ajk >1
,then the jth criterion is more important than the kth
criterion, while if ajk<1,then the jth criterion is less
important than the kth criterion, if two criteria have
the same importance, then the entry ajk is 1.The
Page 94
mess
age
Infor
matio
n
Leaka
ge
ation
gatheri
ng
tion
Disclosu
re
messages
within
SOAP
faults can
contain
detailed
platform
informatio
n and
implement
ation
details
such as
code
fragments
or stack
traces.
g
4
Nume
rical
Value
s
Fuzzin
g
Informa
tion
Disclosu
re
Any value
that is only
as a
numerical
value or is
expected
to be a
numerical
value.
Ratin
g
5
Base6
4
Enco
ded
Value
s
Fuzzin
g
Tamperi
ng
Base64 is
used to
encode
binary
data in
order to
conform to
XML
specificati
ons.
Ratin
g
6
Chara
cter
String
s
Fuzzin
g
Tamperi
ng
This
verybroad
category
general
guidelines
for any
data that is
not of any
particularl
y
classifiable
form.
Ratin
g
7
Gener
al
value
s
Fuzzin
g
Tamperi
ng
If it is not
possible to
indentify
the nature
of the
values
beings
supplied
this
category
provides a
general
Ratin
g
entries ajk and akj satisfy the following constraint
[13]
Once the matrix A is built, it is possible to derive from
A the normalized pairwise comparison matrix A norm
by making equal to 1 the sum of the entries on each
column, i.e. each entry jka of the matrix A norm is
computed as
Finally, the criteria weight vector w (that is an mdimensional column vector) is built by averaging the
entries on each row of A norm, i.e.
4. RISK EVALUATION
There are many different approaches to risk analysis.
Our approach presented here is based on these
standard methodologies and is customized for web
services security testing methodology based on
STRIDE attacks. The STRIDE (Spoofing, Tampering,
Repudiation, Information disclosure, Denial of
Service, and Elevation of privilege) model was used
by Microsoft for categorizing threats (Casteele, 2005).
This means that the STRIDE model is used to
categorize the threats by taking into account their
effects on the security of the application
(Casteele,2005)[14].
Table 2. Threat Mappings
Sl
.N
o
Para
mete
r
Asset
1
Wsdl
Scann
ing
Inform
ation
gatheri
ng
Informa
tion
Disclosu
re
It
describing
the
functionali
ty offered
by the web
service
and the
parameter
s required
to use it.
Ratin
g
2
Web
Meth
od
Enum
eratio
n
Inform
ation
gatheri
ng
Informa
tion
Disclosu
re
Not all
implement
ed
methods
may be
published
in the
WSDL
document
Ratin
g
3
Error
Inform
Informa
Error
Ratin
Threat
© 2014, IJOURNALS All Rights Reserved
Descriptio
n
Meas
urem
ent
Page 95
overview
of the
types of
inputs that
should be
tested.
8
Sub
syste
m
para
meter
Fuzzin
g
Spoofin
g
This
category
relates to
any values
that may
used to
influence
output on
the client
side of the
application
.
Ratin
g
9
Addr
essin
g
para
meter
s
Fuzzin
g
Tamperi
ng
System
often use
addressing
informatio
n to access
informatio
n
directories
.
Ratin
g
Loggi
ng
value
s
Fuzzin
g
Any value
that is
logged
directly to
some
medium
has the
potential
to
somehow
corrupt
logs or
provide an
inaccurate
view.
Ratin
g
Any value
that may
be used as
part of an
SQL query
should be
tested for
the ability
to change
SQL
processing
in some
way,possib
ly causing
data
disclosure.
Ratin
g
If an
Ratin
1
0
1
1
1
Sql
Inject
ion
Com
Injecti
on
Injecti
Tamperi
ng
Spoofin
g
Tamperi
© 2014, IJOURNALS All Rights Reserved
2
mand
Inject
ion
on
ng
internal
system is
used to
execute
existing
commands
and input
to these
commands
is not
properly
validated,it
may be
possible to
run
commands
of the
user’s
choosing.
g
1
3
Lpath
Injecti
on
Spoofin
g
If LDAP
queries are
constructe
d directly
from user
input, this
may result
in
significant
system
compromi
se,particul
arly in the
disclosure
of user
credentials
.
Ratin
g
Inject
ion
1
4
Xpath
Inject
ion
Injecti
on
Informa
tion
Disclosu
re
The use of
user
supplied
input in an
XPath
query may
provide an
attacker
with the
ability to
modify the
query.
Ratin
g
1
5
Code
Injecti
on
Elevatio
n of
Privileg
es
If an
validated
user
supplied
input is
supplied to
calls to
eval-type
functions,
malicious
commands
may be
Ratin
g
Inject
ion
Page 96
inadverten
tly
executed
by the web
service
1
6
Ciphe
r
Confid
entialit
y
choic
e
Informa
tion
Disclosu
re
The choice
of
encryption
cipher will
influence
the
strength of
the
encryption
and the
ability for
an attacker
to
successfull
y crack the
encryption
and
recover
plaintext
data
Ratin
g
1
7
Encry
ption
Cover
age
Confid
entialit
y
Informa
tion
Disclosu
re
Encryption
should be
applied
overall
sensitive
portions of
messages
to ensure
they are
protected
against un
authorized
eaves
dropping..
Ratin
g
1
8
Repla
y
Attac
ks
Integri
ty
Spoofin
g
A replay
attack
involves
the
malicious
use of a
valid
message or
set of
messages
that has
already
been
accepted
by the web
service
previously.
Ratin
g
Integrity
checks
should be
used to
Ratin
g
1
9
Integ
rity
Check
Cover
Integri
ty
Tamperi
ng
© 2014, IJOURNALS All Rights Reserved
protect
important
data
against
unauthoriz
ed
modificati
on.
age
2
0
Invali
d Xml
Integri
ty
Denial
of
service
WSSecurity
and other
web sevice
security
standards
are XMLbased and
their
implement
ations
require
properly
formed
XML to
function
properly.
Ratin
g
2
1
Unsu
pport
ed
algori
thms
Integri
ty
Tamperi
ng
Verify that
if
unsupport
ed
algorithms
are
requested
or the
client
claims to
root
support
required
algorithms,
access is
denied and
processing
of the
request
does not
continue.
Ratin
g
2
2
Separ
ator
Inject
ion
Loggin
g
Repudia
tion
Log entries
are
commonly
delimitedu
sing a
particular
separator
character.
Ratin
g
2
3
White
Space
Inject
ion
Loggin
g
Repudia
tion
White
space
characters
can be
used to
modify the
Ratin
g
Page 97
appearanc
e of log
entries
when they
are
viewed.
2
4
Brute
-
Authen
ticatio
n
Force
and
Dictio
nary
Attac
ks
2
5
2
6
2
7
Forge
d
Crede
ntials
Missi
ng
Crede
ntials
Toke
n
Forge
ry
Authen
ticatio
n
Authen
ticatio
n
Author
ization
Elevatio
n of
Privileg
es
Elevatio
n of
Privileg
es
Spoofin
g
Elevatio
n of
privileg
es
© 2014, IJOURNALS All Rights Reserved
These
types of
attacks are
typically
used
against
password
authentica
tion
systems
and rely on
the ability
to
repeatedly
test
potential
passwords
against the
authentica
tion
service.
Ratin
g
Credential
s should be
issued by
an
authorized
party and
verified by
the
application
when
presented.
Ratin
g
A user that
fails to
present
credentials
should not
be allowed
access and
the
application
should
discard
their
request.
Ratin
g
As SOAP is
a stateless
messagebased
protocol
,some
mechanis
m must be
implement
Ratin
g
ed to
provide
authorizati
on
between
SOAP
requests
or
maintain
session
state.
2
8
Hijac
king
Attac
ks
Author
ization
Tamperi
ng
As SOAP is
a stateless
messagebased
protocol,
some
mechanis
m must be
implement
ed to
provide
authorizati
on
between
SOAP
requests
or
maintain
session
state.
Ratin
g
2
9
Para
meter
Tamp
ering
Availa
bility
Denial
of
service
This broad
class of
attacks
refers to
the
modificati
on of SOAP
request
parameter
s in transit
between
client and
server.
Ratin
g
3
0
Coerc
ive
Availa
bility
Denial
of
service
Coercive
parsing is
the attacks
that
involve
supplying
illegal or
malformed
SOAP
requests to
the web
service in
order to
cause
undesirabl
Ratin
g
Parsi
ng
Page 98
e behavior.
S1=0.22
S2=0.13 S3=0.1 S4=0.25
T1=0.28 T2=0.12 T3=0.22 T4=0.91 T5=0.23
SOAP Sonar web services penetration testing tool
used to evaluate web services security under known
attacks [15]. Bringing this approach to find attacks
that found the major security flaws of the web service
meaning the four Spoofing, five Tampering and four
Reputation Parameters attacks errors that poses
number of threat to web service security.
If we involve all the thirty parameters mentioned in
table2, complexity for estimating the effectiveness of
web service increases. As a solution to this, we follow
the mechanism of greedy approach. The core
principle of the Greedy approach is to involve
minimum number of input to achieve the final
optimized result.
Spoofing Attacks - Sub-system, SQL Injection, Xpath
Injection, Replay Attacks.
Tampering Attacks - Code Injection, Invalid XML,
Encryption Coverage, Numerical Values, Command
Injection.
Repudiation Attacks - White space injection,
Separator Injection, HTML Injection, Size Overflow.
We do pair wise comparing between criteria and the
below matrix is the result of the comparison with
respect to Table1.We assume that based on above
attacks exits in number of times with respect of four
QoP criteria is gathered and the matrix is prepared.
Table 3. Preference Level for Spoofing
R1=0.37 R2=0.15 R3=0.28 R4=0.20
We assume that based on above example the data of
six alternatives STRIDE with respect of four QoP
criteria is gathered and the overall matrix is prepared
with respect to number of attacks existing in each
web service using Soap Sonar test tool.
5. CONCLUSIONS
In this paper we addressed web service security by
involving Greedy approach to select desired security
parameter using stride pattern and AHP method for
evaluation weighting of criteria for each web service
for final ranking.
6. REFERENCES
[1] Bin Xu, Tao Li, Zhifeng Gu, Gang Wu “Quick Web
Service Discovery and Composition in SEWSIP”,
Proceedings of the 8th IEEE International
Conference on E-Commerce Technology and the
3rd IEEE International Conference on Enterprise
Computing, E-Commerce, and E-Services
(CEC/EEE’06).
[2] J. Viega and G. McGraw, Building Secure
Software, How to Avoid Security Problems the
Right Way, Addison Wesley, 2002
[3] G. Hoglund and G. McGraw, Exploiting Software,
How to Break Code, Addison Wesley, 2004.
[4] M. Howard and D. LeBlanc, Writing Secure Code,
Microsoft Press, 2002.
[5] Vu, L., Hauswirth, M., and Aberer, K. (2005).
“QoSbased service selection and ranking with
trust and reputation management”. In Proc. of
the Intl. conf. on Cooperative Information
Systems (CoopIS), Agia apa, Cyprus.
Table 4. Preference Level for Tampering
[6] J.Scambray and M.Shema, Hacking Exposed Web
Applications, McGrawHill, 2002
[7] Li Jiang, Hao Chen, Fei Deng, Qiusheng Zhong.“A
Security Evaluation Method Based on Threat
Classification for Web Service”, Journal of
Software, Vol 6, No 4 (2011), 595-603, Apr 2011
Table 5. Preference Level for Reputation
[8] Bachar Alrouh and Gheorghita Ghinea. “A
Performance Evaluation of Security Mechanisms
for Web services”, Fifth International Conference
on Information Assurance and Security 2009.
[9]
Based on the above table we apply steps1-3 in section
3.3 and the weights of criteria are as follow:
© 2014, IJOURNALS All Rights Reserved
Artsiom Yautsiukhin. “Quality of Protection
Determination for Web Services”. This work was
partly supported by the project EU-IST-IPSERENITY, contract N 27587.
[10] GannaFrankova
and
ArtsiomYautsiukhinork.“Service and Protection
Level Agreements for Business Processes”, This
work has been partly supported by the IST-FP6IP-SERENITY and IST-FP6-IP-SENSORIA projects
Page 99
[11] González, Ricardo Mendoza. “Web ServiceSecurity Specification based on Usability Criteria
and Pattern Approach” Journal of Computers .
Aug2009, Vol. 4 Issue 8, p705-712. 8p
[12] Duan Youxiang, Gao Yang. “Evaluating
Vulnerabilities Quantitatively Based On the Rank
of Web Services Confidentiality” Journal of Next
Generation Information Technology, volume 2,
Number 1, February, 2011
[15] http://www.softpedia.com/get/Authoringtools/Authoring-Related/SOAPSonar-EnterpriseEdition.shtml
[16] Mojtaba Khezrian1, Wan M. N. Wan Kadir2,
Suhaimi Ibrahim3,‎ and‎ Alaeddin‎ Kalantari.‎ “A
Hybrid Approach for Web Service Selection”,
/International
Journal
Of
Computational
Engineering Research Jan-Feb 2012 , Vol. 2,
Issue1.
[13] Analytic Hierarchy Process,
www.dii.unisi.it/~mocenni/Note_AHP.pdf‎
[14] Casteele, S.V. (2005). Threat modeling for web
application using STRIDE model.
© 2014, IJOURNALS All Rights Reserved
Page 100

Download Report